class: center, middle, inverse, title-slide .title[ # Class 1: Introduction to Business Forecasting ] .author[ ### Krzysztof Zaremba ] --- <style type="text/css"> .remark-slide-content { font-size: 20px; } </style> ## Who am I? - **Krzysztof Zaremba** - **Education:** PhD in Economics, Columbia University - **Research Focus:** Applied Econometrics, Health Economics - **Email:** .blue[zaremba@itam.mx] - **Office:** Santa Teresa Campus - **Office Hours:** Mondays 2–3pm on Zoom (email for other times) --- ### What Is This Class About? **Forecasting** is the science and art of making predictions about future events using historical data and relevant information. -- In this class, you will learn how to turn data into actionable insights that drive better decisions in business and beyond. --- #### Why is Forecasting Important? -- 1. **Weather Forecasting** - Predicting hurricanes → timely reactions and preparations. <center> <img src="data:image/png;base64,#https://ral.ucar.edu/sites/default/files/public/images/features/hfip1.png" /> </center> --- 2. **Epidemic Forecasting** - Predicting disease outbreaks → effective public health responses. <center> <img src="https://www.washingtonpost.com/wp-apps/imrs.php?src=https://arc-anglerfish-washpost-prod-washpost.s3.amazonaws.com/public/RA3SMOZBVRAKREEJ7HBLLRITEI.png&w=1440" width="600"> </center> --- 3. **Crime Forecasting** - Identifying crime hot spots → more effective police patrols. - Identifying individuals at risk → targeted interventions. <center> <img src="Crime_prediction.png" width="600"> </center> --- One of your tasks later in class: **Predict 911 calls in Mexico City** -- <center> <img src="choropleth_investigations_quantiles.png" width="600"> </center> --- ### What Is This Class About? #### Forecasting in Business - With the increasing availability of data, firms rely on rigorous data-driven methods to make decisions. - Many business decisions depend on anticipating future events. - In this course, you’ll learn how to use data to make informed predictions about the future. --- ### Applications of Forecasting 1. **Anticipating Demand** Adjust inventory to minimize waste and meet customer needs effectively _Example: Meal kits, retail chains_ 2. **Employee Churn** Predict which employees are likely to leave and develop strategies to retain talent _Example: HR analytics in consulting firms_ 3. **Strategic Investment Decisions** Use market forecasts to guide investment timing and location _Example: Expansion of electric vehicle infrastructure_ 4. **Improving Advertisement Effectiveness** Identify which components of an ad drive conversions _Example: A/B testing by Airbnb_ 5. **Product Recommendations** Forecast customer preferences based on past behavior _Example: Amazon, Netflix recommendation systems_ --- #### Learning to Predict Real World Scenarios: Predicting Bus Ridership in NYC **Final Project Context** - The NYC Department of Transportation is forecasting number of customers to optimize resources: - How many buses are needed on a given day? - How many drivers should be scheduled? **Project Goal** - Predict the **number of passengers using buses in NYC on December 3rd** (after the submission day). **Data Insights** - Rich quantitative data: Each payment generates a datapoint on customer bus usage. - External factors affecting demand include: - School and holiday calendars - Weather conditions - City events - Road repairs --- <!-- --> --- <!-- --> #### **Key Outcome** - Students built a **linear regression model** combining multiple data sources to generate forecasts. - **Closest prediction was within 1% of the actual number of riders!** - A reliable tool to improve resource allocation and decision-making. --- ### Your turn - Get in pairs - Consider your past employment or your future employment - Think about how forecasting could solve some problems in the context of industry you are considering - (5 min) --- ### What will you learn? 1. Getting Business Information from the Data - Analyze data, evidence, and arguments to make reasoned judgments -- 2. Problem Solving and Forecasting - Formulate, evaluate, and implement statistical models for business forecasting. - Interpret the results and validate assumptions - Key technical skill very valuable on the job market! -- 3. Decision Making and Communication - Choose optimal options to achieve objectives. - Communicate findings, conclusions, and recommendations effectively to business professionals --- ### Organization - **Lectures Schedule**: - Tuesday morning RH SA2 - **TA schedule**: - Weekly 1 hour lab session - timing/location to be determined - Office Hours with TAs - timing/location to be determined - Canvas Discussion Board - **Textbooks** - See canvas course materials and syllabus - **Organization** - see the website --- ### Grading -- - **70% In-Class Exams & Assignments** - 3 total exams (including final) - Combination of applied (in computer lab) and paper-based formats -- - **20% Projects & Presentations** - 3 group-based take-home projects - Analyze data, write a short summary - Present findings briefly in class -- - **5% In-Class Assignments** - 2 group forecasting competitions using real-world data - 2 individual theoretical quizzes (1 cheat sheet allowed) - **Lowest score among these 4 assignments will be dropped** -- - **5% Participation & Engagement** - 2.5%: Flipped tasks and short homeworks (graded for completion) - 2.5%: Preparedness checks at the beginning of class - Random students selected in the first 5 minutes - Zero credit if absent or unprepared --- ### Pre-requisites - Mathematics III or Linear Algebra I - Statistics II or Statistical Inference --- ### Language - Class is in English -- - But your English is not evaluated -- - If you don't know a word, feel free to ask -- - Great occasion to learn vocabulary useful for interviews --- ### ChatGPT & AI Tools Policy -- - **You are allowed to use ChatGPT and other language models.** - Especially useful for refining code, summarizing concepts, and improving writing in projects. -- - **But be careful — copying output directly is plagiarism.** - You must **understand and be able to explain** anything you include in your work. - During presentations, you will be expected to explain your project. If you can't, you’ll receive **zero credit** for that assignment. -- - **ChatGPT is not allowed during exams.** -- - **Use AI tools wisely — to learn, not to outsource thinking.** -- - **Recommended for learning and review. Try asking:** - "Why can’t we say a 95% confidence interval contains the true value with 95% probability?" - "What’s the relationship between Type I and Type II errors?" - "What is a p-value and what is it not?" --- <center> <img src=Chatgptlearning.png width="800"> </center> --- ### Academic Integrity -- - It disrespects those supporting your education — you are not learning anything. - It is unfair to your peers who put in genuine effort. - It disrespects the professor and the academic environment. -- - **Cheating is easy to detect**: - Quiz and exam questions are randomized. - A LockDown Browser will be used during quizzes. - Accessing quizzes outside the classroom without prior approval = **0 credit**. - Canvas access logs will be reviewed during assessments. - In-class assignments are recorded. - Anonymous reporting of cheating is available. -- - **Penalties are severe**: - If caught cheating: - You will be reported to the administration. - You will automatically fail the course and must retake it. - On a second offense, **you will be expelled**. - Cheating has been reported in this course before. -- - If you act with honesty and integrity, **you have nothing to worry about**. --- ### Software: R **Performing Real Forecasts with Real Data** -- - **Widely Used** R is one of the most popular languages for data analysis, statistical modeling, and forecasting — especially in academia and industry. -- - **Free and Open-Source** R is completely free to use and supported by a large global community. -- - **Great Community Support** Thousands of tutorials, packages, and forums are available — it’s easy to find help when you're stuck. -- - **Powerful Data Visualization** R includes powerful tools (like `ggplot2`) to create clear, publication-quality visualizations that help you communicate your insights effectively. --- ### Life expectation vs GDP per Capita
--- ### Bitcoin price in time
--- ### Remarks - Knowledge of theoretical statistics does not matter much if you can't apply it using modern tools - We will use it for practical exercises with data - You will use it for the final project - You will use it for the applied exams - We will learn some of it together in class - TA lab sessions will further help with this - Chatgpt is your friend --- ### Introduction to Forecasting Tools will often depend on the horizon and data availability -- #### Forecasting Horizons - .blue[Very Short Horizon:] - High-Frequency Trading: Real-time price predictions for financial trading -- - .blue[Short Horizon]: - Public bikes: Predict the availability of bikes at bike station and adjust the number -- - .blue[Long Horizon]: - New obesity drug: Forecasting number of potential patients and their resources --- ### Overview of Forecasting Techniques 1. Qualitative Forecasting - Based on subjective judgment and expert opinions - Suitable for unique situations or new markets - Examples: Predicting economic impacts of oil price changes or political stability in a region -- 2. Quantitative Forecasting - Uses historical data and numerical techniques - Suitable when data is available and continuity assumptions hold true - *Continuity assumption*: past trends and relationships continue in the future - When it holds? -- - Interest rates and investments - When it does not hold? -- - Covid Cases & Deaths and Vaccines --- layout: false class: inverse, middle # Methods of Quantitative Forecasting --- ## Time Series Forecasting .pull-left[ - .blue[Time series data]: collection of data points for a single unit (one firm, one person, one country) ordered chronologically. Can be one or more variables. - .blue[Time series forecasting]: identifying patterns and trends in historical data to predict future values ] .pull-right[
] -- #### In simple terms: - We don't care about what causes what - We just hope that past values of the variable and its historical behavior can predict its future values --- ## Example: Forecasting of GDP - Time series forecasting can help predict a country's Gross Domestic Product.  --- ## Explanatory Models - We have data on both the variable of interest and other variables related to it - We consider how other variables impact the outcome of interest - We use these relationships to make forecasts --- .pull-left[## Example: Sales at a new location - Should we open a new Starbucks at ITAM? - Using existing locations, analyze impact of: - Foot traffic - Neighborhood income - Competitors' stores - Given these relationships, what would be sales at ITAM? ] .pull-right[ .white[a]
] --- # Steps of Forecasting 1. **Problem Definition** - Clearly define the forecasting objective. 2. **Gather Data** - Identify and collect relevant data. 3. **Preliminary Explanatory Analysis** - Understand data characteristics and relationships. 4. **Choosing and Fitting the Model** - Select and fit the appropriate forecasting model. 5. **Evaluating the Model** - Assess the model's performance using historical data. 6. **Communicate the results** - Visualize and interpret the outcomes and write a report or presentation --- <!-- layout: false --> <!-- class: inverse, middle --> <!-- # Methods of Qualitative Forecasting --> <!-- --- --> <!-- <center> --> <!-- <img src=Delphi.jpg width="600"> --> <!-- </center> --> <!-- --- --> <!-- ### Delphi Method --> <!-- - A structured communication process to reach a consensus for complex, uncertain and long terms forecasting tasks --> <!-- 1. Select a group of experts --> <!-- 2. Invite them to the study. They are anonymous and don't talk to each other! --> <!-- 3. Ask them to answer a questionnaire --> <!-- 4. Get initial responses --> <!-- 5. Compile them into summary --> <!-- 6. Send them summary and get their feedback with refined answers --> <!-- 7. Reiterate until consensus is reached or no further improvement --> <!-- -- --> <!-- #### Example: Determining AI threats --> <!-- - What are the risks of AI developments? --> <!-- - Panel of experts from academia and industry --> <!-- - Computer scientists, engineers, CEOs of AI companies, ethic experts --> <!-- - Send them questionnaires asking about potential threats --> <!-- - Compile responses into summary and send them back --> <!-- - Get more rounds of responses until consensus --> <!-- - Identify the most probable risks --> <!-- --- --> <!-- ### Brainstorming --> <!-- - Creative technique for generating ideas. --> <!-- - Encourages free thinking and building on suggestions. --> <!-- - Appropriate for exploring possibilities. --> <!-- - Form a group (no need for experts) --> <!-- - State the problem --> <!-- - Encourage ideas, no matter how crazy --> <!-- - Build and combine each others' ideas --> <!-- - Document the ideas and synthesize them --> <!-- -- --> <!-- #### Example: Enhancing Employee Engagement --> <!-- - Tech company's HR department. --> <!-- - Representatives from HR, IT, and different departments. --> <!-- - Generate ideas for a mobile app to enhance employee engagement. --> <!-- - Write them down and implement the relevant ones --> <!-- --- --> <!-- ### Panel of Experts --> <!-- - Assemble knowledgeable individuals --> <!-- - At the same time and spot --> <!-- - They meet, offer insights and expertise, and discuss --> <!-- - Aid in well-informed decisions. --> <!-- - Sometimes ends up with a report with conculsions --> <!-- -- --> <!-- #### Example: Environmental Policy Formulation --> <!-- - Government agency want to find identify and address most pressing environmental issues --> <!-- - Environmental scientists, economists, conservationists, and policymakers. --> <!-- - Discuss policy options. --> <!-- - Create comprehensive environmental policies. --> <!-- --- --> <!-- ### Focus Groups --> <!-- - Gather diverse participant - not necessarily experts --> <!-- - Share perceptions, attitudes, and opinions. --> <!-- - Provide qualitative data and consumer insights. --> <!-- -- --> <!-- #### Example: Market Research for a New TV SHOW --> <!-- - Proposing a new TV Show and trying to see how well it will do --> <!-- - Participants from various demographics. --> <!-- - Understand consumers' preferences and perceptions about the TV show --> <!-- - Fine-tune the product and marketing strategy. --> <!-- --- --> ### Remainder of the course #### Quantatitve Forecasting 0. Ungraded quiz -- 1. Review of Statistics -- 2. Simple linear regression -- 3. Multiple linear regression -- 4. Time Series --- # Questions?